Literacy is closely linked with health status and health care utilization. For Medicaid enrollees, the average reading level is grade five; as more and more states turn to mandatory managed care in their Medicaid programs, these enrollees may find it difficult to navigate an increasingly complex health care system. One recognized measure of quality of care is patient satisfaction-a measure that must be collected from patients themselves. Patients with low literacy skills may face written, self-administered questionnaires that are beyond their ability to comprehend-thereby systematically excluding these stakeholders from this important quality measure. One instrument holds promise for assessing satisfaction in this vulnerable group: the Consumer Assessment of Health Plans Study (CAHPSTM) survey. It is written at a fifth-grade level, and has been tested in the Medicaid population. In this project, we will develop and test two alternative formats for the CAHPSTM survey to make it even more accessible to low-literacy populations. Specific study goals include: 1) to develop an illustrated form of the CAHPSTM 2.OH self-administered instrument; 2) to develop an interactive voice response (IVR) version of CAHPSTM 2.0H for telephone self interview; 3) to test the acceptability and psychometric properties of these CAHPSTM formats in a Medicaid managed care population; and 4) to compare, in a randomized controlled trial, alternatives for administering the survey: the existing CAHPSTM, an illustrated version, an IVR version, and the existing CAHPSTM with an IVR option. The study population will comprise adults who are enrolled in Health Partners, a Medicaid managed care organization in Philadelphia. The racial distribution of enrollees is 51 percent African American, 23 percent Hispanic, 16 percent White, 4 percent Asian American, and 6 percent other. With feedback from focus groups of Health Partners' members, we will develop and test the prototypes for the illustrated and IVR versions. Then, we will test the instruments for acceptability, validity and reliability in multiple clinic sites affiliated with Health Partners, and assess literacy levels of the respondents to determine how well the instruments perform across literacy groups. Finally, we will randomize eligible enrollees to receive the standard form or one of the alternative versions, to determine how the instruments perform in the typical way in which they would be used.